摘要
针对移动互联网QoS优化中存在的测量不精确的问题,提出了异构融合移动互联网面向多种业务非精确参数的QoS路由算法。多目标优化模型是本文算法的基本模型,在业务感知的条件下实现约束的输入,在多量纲的基础上尽可能实现多目标的优化并利用覆盖网路由来保障QoS。为解决数据不精确的问题,结合概率方法和隶属度对多次采样QoS数据进行模糊处理,并构建方程利用理想点以解决网络性能均衡等问题。通过随机拓扑进行仿真,在满足业务传送约束的基础上,与同类型算法相比,该算法兼顾链路和节点的性能,在同样的网络环境下提升了QoS满意率,降低了时间开销,并能有效提高系统的资源利用率。
An overlay quantity-of-service （QoS） routing algorithm is proposed for a vast variety of services because the QoS optimization in heterogeneous and converged mobile-internet usually has inaccurate measurements. The algorithm is based on the model of multi-objective optimization and the service-aware requirements are inputs of the model. The multiple objects are optimized well and the QoS is ensured using the overlay QoS routing. The algorithm deals with the inaccurate measurements by building an equation to fair the network resource as far as possible. Multi QoS parameters are collected by sampling, and membership functions and probability theory are combined to fuzzy these parameters. Then multi-objective optimization algorithm is employed to get a solution set. A series of simulation results with random topology has confirmed the advantage of the scheme. Comparisons with other traditional algorithms show that the proposed algorithm improves the service QoS satisfaction rate, reduces the run time-overhead and enhances the resource utilization with taking the performance of nodes and links into account.
出处
《西安交通大学学报》
EI
CAS
CSCD
北大核心
2014年第2期86-92,共7页
Journal of Xi'an Jiaotong University
基金
国家自然科学基金资助项目(61071126)
国家科技重大专项资助项目(2010ZX03004-002-01
2011ZX03002-001-02
2013ZX03002010-003)
关键词
移动互联网
QOS路由
多目标优化
业务感知
非精确参数
mobile-internet
QoS routing
multi-objective optimization
service-aware
inaccurateparameters
